Fecal incontinence (FI) affects 9% of U.S. adults and occurs weekly or more often in 2.7%. Prevalence increases with age reaching 15% by 70 years. FI has a devastating impact on quality of life and substantially increases caregiver burden when patients have comorbid conditions requiring caregiver assistance. We propose an effectiveness trial whose overall goal is to determine whether a conservative intervention which has been shown to be effective in single-site studies will sustain its efficacy when disseminated to a home bound population by home health care nurses. The treatment includes patient education about the physiology of how continence is maintained, pelvic floor exercises, behavioral strategies for preventing FI, and use of fiber or nonprescription medication to treat diarrhea or constipation. To minimize drift when the treatment is disseminated to a large group of providers, patient education and other basic components of treatment are included in a DVD; however, nurse supervision to individualize treatment remains important. The study will be carried out in 8 counties in central North Carolina which are served by the University of North Carolina Home Health Care and Rex Home Health Care agencies (partners in this study). These counties will be randomly assigned to two groups approximately matched on urban vs. rural residence. For the first two years all patients in one group of counties will receive the active intervention while residents of the other counties receive usual care plus a DVD and symptom monitoring to control for expectancy. After two years, patients in all 8 counties will receive the active intervention. After excluding patients with severe cognitive impairment and those with stomas, we estimate 252 - 340 patients with FI who are over age 50 will receive treatment (at least 189 in active treatment and 63 controls). Specific aims are (1) to show that the active treatment is more effective than the control treatment for improving FI severity, patient quality of life, and caregiver burden, and that improvements are maintained for at least 6 months; (2) to identify moderators of treatment effectiveness (candidate variables are cognitive status, mobility impairment, willingness of family caregiver to assist with treatment, anxiety, depression, age, and race); and (3) to explore whether successful treatment of FI reduces the risk of nursing home referral. This is a revised application.